There is an urgent public health need to develop effective pharmacotherapies for methamphetamine (meth) dependence. Globally, use of amphetamines is more common than heroin and opioids, with an estimated 16 to 51 million users worldwide and 1.3 million users in the U.S. in 2007.1 Use of meth (rapidly metabolized in the bloodstream to amphetamine) is up to 20 times more prevalent among men who have sex with men (MSM) than in the general U.S. population, and is a major contributor to the continued HIV epidemic. 2-4 Meth use is independently associated with high-risk sexual behavior and HIV seroconversion.5-7 Effective meth treatments therefore will not only reduce meth use, they may also serve as important HIV prevention interventions by reducing meth-driven sexual risk behavior. The lack of FDA-approved pharmacotherapies for meth dependence has limited treatment to behavioral interventions which may benefit from additional adjunctive pharmacologic treatment.8, 9 Furthermore, many meth users, including MSM, do not seek conventional drug treatment, reinforcing the need to develop new treatment modalities.10 Naltrexone, a 5-opioid receptor antagonist, is a highly promising agent that has shown efficacy in reducing relapse of amphetamine use among amphetamine-dependent, yet currently amphetamine- abstinent heterosexuals.11 We propose to expand upon this promising work to determine whether naltrexone will reduce methamphetamine use among actively using, meth-dependent MSM with high-risk sexual behavior through a double-blinded randomized controlled trial of naltrexone versus placebo. We will focus on MSM because of the disproportionate and intertwining epidemics of meth use and HIV in this population. The primary specific aims of this study are: 1) To determine the efficacy of oral naltrexone versus placebo in reducing meth use among actively-using, meth-dependent MSM with high-risk sexual behavior, as determined by the proportion of meth- metabolite positive urines by study arm. 2) To determine the efficacy of oral naltrexone versus placebo in reducing meth-associated sexual risk behavior as measured by: numbers of male anal sex partners, serodiscordant unprotected anal sex partners, and numbers of sex partners with whom meth was used, by study arm. 3) To evaluate the safety and tolerability of oral naltrexone versus placebo, as determined by adverse clinical event rates by study arm. 4) To determine the acceptability of oral naltrexone versus placebo by evaluating adherence via medication event monitoring systems (MEMS) caps, urine ss-naltrexol, urine riboflavin, and self-report. We will enroll 100 sexually active, meth-dependent MSM who will be randomized 1:1 to receive naltrexone or placebo for 12 weeks. All participants will receive substance use counseling and HIV risk-reduction counseling.